An adaptive controller for photovoltaic emulator using artificial neural network

Abstract

The photovoltaic (PV) emulator is a nonlinear power supply that features the similar characteristic of the PV module. However, the nonlinear characteristic of the PV module causes instability of the PV emulator output. The conventional solution is to operate the PV emulator in the overdamped condition which results in a poor dynamic performance. This drawback is solved by manipulating the proportional and integral gains of the proportional-integral (PI) controller. In this paper, the artificial neural network is used in the adaptive PI controller to maintain a stable and fast dynamic response of the PV emulator. This has been simulated with varied output resistance and irradiance. By comparing the proposed control strategy with the conventional method during start-up response of the photovoltaic emulator, the dynamic performance of the output current has shown an improvement of up to 80 % faster than the conventional method.